Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendment
The Amendment filed 03/16/2026 has been entered. Claim 10 is canceled and claims 13-14 are new. Claims 1-9 and 11-14 are pending in this application.
Allowable Subject Matter
Claims 4-5 and 9 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
Specification
The title of the invention is not descriptive. A new title is required that is clearly indicative of the invention to which the claims are directed. The following title is suggested: “Time-Series Diagnosis Using Multidimensional Probabilistic Model Matching”
Claim Objections
Claim 1 is objected to because of the following informalities:
Claim 1, line 2 recites the phrase “configured to” which should be “configured to:”.
Claim 1, line 4+ ensure consistency between a first probabilistic model, the plurality of models and a probability model. What types of models are there and which type does the first belong.
For the informalities above and wherever else they may occur appropriate correction is required.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102 of this title, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1, 8, 11-12 and 14 are rejected under 35 U.S.C. 103 as being unpatentable over Geipel et al. (US 20220147034 A1 hereinafter Geipel) in view of Wren et al. (US 7542949 B2 hereinafter Wren) and THIRUVENKATANATHAN et al. (US 20240353825 A1 hereinafter Thiru)
As to independent claim 1, Geipel teaches an information processing device, comprising: [device 600 ¶56]
one or more hardware processors configured to [processor 610 ¶56]
a probability model corresponding to each dimension of each of the plurality of multidimensional probability distribution models being a model that models probabilities of a plurality of values, at a corresponding time, of time series data whose data length is L, the data length of the plurality of pieces of partial time series data being L, [Models assign probability to time series data at interval (length)¶37 "This first and second probabilistic model 225, 235 may each be provided in the form of a function which, for a certain data point or time interval of the time series data 20, outputs a probability of observing this data point time interval. "]
the plurality of pieces of partial time series data being contained in target time series data to be a target of diagnosis; and [device describes time series data dynamics (its diagnosed) in windows (targets) ¶8 "model describing dynamics of the time series data inside the first time window"]
determine a plurality of pieces of matching information including positions of the plurality of pieces of partial time series data in the target time series data, [determines matches ¶8 " determine a first part of the time series data that is estimated to match the first probabilistic model and a second part of the time series data that is estimated to match the second probabilistic model"]
Geipel does not specifically teach calculate first similarities between a plurality of multidimensional probability distribution models and a first probabilistic model whose first similarity with respect to the plurality of pieces of partial time series data at the positions is maximum, and the first similarity to the first probabilistic model.
However, Wren teaches calculate first similarities between a plurality of multidimensional probability distribution models and [measures similarities (likelihood) between model observations Col. 3 ln. 27-38 "if two observations are similar, then the two models constructed and trained for the two observations are similar. That is, the model for one observation generates the other observation with a high likelihood "]
a first probabilistic model whose first similarity with respect to the plurality of pieces of partial time series data at the positions is maximum, and the first similarity to the first probabilistic model. [maximum likelihood (larger) used to determine models Fig. 1 200 among those in the tree 100 Col. 2 ln. 45-52 "Each data sequence is assigned to only a single subtree, specifically, the subtree associated with the single-path model that generates the data sequence with a maximum likelihood. In this way, the single-path models define the patterns in temporal data sequences. The method used to learn the path models from the temporal data sequences is described in detail below."]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the models disclosed by Geipel by incorporating the calculate first similarities between a plurality of multidimensional probability distribution models and a first probabilistic model whose first similarity with respect to the plurality of pieces of partial time series data at the positions is maximum, and the first similarity to the first probabilistic model disclosed by Wren because both techniques address the same field of time series data and by incorporating Wren into Geipel provides better discovering of temporal pattern in data using less time and resources [Wren Col. 1 ln. 42-53]
Geipel and Wren do not specifically teach a plurality of pieces of partial time series data whose data length is L, a number of dimensions of each of the plurality of multidimensional probability distribution models being L,
However, Thiru teaches a plurality of pieces of partial time series data whose data length is L, a number of dimensions of each of the plurality of multidimensional probability distribution models being L, [model dimensions according to components (length) of time-series data ¶38 "anomaly detection envelope or volume having a dimension corresponding to the number of correlated time series data components and/or features"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the models disclosed by Geipel and Wren by incorporating the plurality of pieces of partial time series data whose data length is L, a number of dimensions of each of the plurality of multidimensional probability distribution models being L, disclosed by Thiru because all techniques address the same field of time series data and by incorporating Thiru into Geipel and Wren can make models more accurate and operate faster [Thiru ¶22].
As to dependent claim 8, the rejection of claim 1 is incorporated, Geipel, Wren and Thiru further teach wherein the one or more hardware processors is further configured to train the plurality of multidimensional probability distribution models by using a plurality of pieces of time series data for learning. [Wren training data sequences (pieces) Col. 2 ln. 37-64 "temporal patterns are learned from training data sequences using the composite HMMs"]
As to independent claim 11, Geipel teaches an information processing method executed by an information processing device, the information processing method comprising: [device 600 ¶56]
a probability model corresponding to each dimension of each of the plurality of multidimensional probability distribution models being a model that models probabilities of a plurality of values, at a corresponding time, of time series data whose data length is L, the data length of the plurality of pieces of partial time series data being L, [Models assign probability to time series data at interval (length)¶37 "This first and second probabilistic model 225, 235 may each be provided in the form of a function which, for a certain data point or time interval of the time series data 20, outputs a probability of observing this data point time interval. "]
the plurality of pieces of partial time series data being contained in target time series data to be a target of diagnosis; and [device describes time series data dynamics (its diagnosed) in windows (targets) ¶8 "model describing dynamics of the time series data inside the first time window"]
determine a plurality of pieces of matching information including positions of the plurality of pieces of partial time series data in the target time series data, [determines matches ¶8 " determine a first part of the time series data that is estimated to match the first probabilistic model and a second part of the time series data that is estimated to match the second probabilistic model"]
Geipel does not specifically teach calculate first similarities between a plurality of multidimensional probability distribution models and a first probabilistic model whose first similarity with respect to the plurality of pieces of partial time series data at the positions is maximum, and the first similarity to the first probabilistic model.
However, Wren teaches calculate first similarities between a plurality of multidimensional probability distribution models and [measures similarities (likelihood) between model observations Col. 3 ln. 27-38 "if two observations are similar, then the two models constructed and trained for the two observations are similar. That is, the model for one observation generates the other observation with a high likelihood "]
a first probabilistic model whose first similarity with respect to the plurality of pieces of partial time series data at the positions is maximum, and the first similarity to the first probabilistic model. [maximum likelihood (larger) used to determine models Fig. 1 200 among those in the tree 100 Col. 2 ln. 45-52 "Each data sequence is assigned to only a single subtree, specifically, the subtree associated with the single-path model that generates the data sequence with a maximum likelihood. In this way, the single-path models define the patterns in temporal data sequences. The method used to learn the path models from the temporal data sequences is described in detail below."]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the models disclosed by Geipel by incorporating the calculate first similarities between a plurality of multidimensional probability distribution models and a first probabilistic model whose first similarity with respect to the plurality of pieces of partial time series data at the positions is maximum, and the first similarity to the first probabilistic model disclosed by Wren because both techniques address the same field of time series data and by incorporating Wren into Geipel provides better discovering of temporal pattern in data using less time and resources [Wren Col. 1 ln. 42-53]
Geipel and Wren do not specifically teach a plurality of pieces of partial time series data whose data length is L, a number of dimensions of each of the plurality of multidimensional probability distribution models being L,
However, Thiru teaches a plurality of pieces of partial time series data whose data length is L, a number of dimensions of each of the plurality of multidimensional probability distribution models being L, [model dimensions according to components (length) of time-series data ¶38 "anomaly detection envelope or volume having a dimension corresponding to the number of correlated time series data components and/or features"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the models disclosed by Geipel and Wren by incorporating the plurality of pieces of partial time series data whose data length is L, a number of dimensions of each of the plurality of multidimensional probability distribution models being L, disclosed by Thiru because all techniques address the same field of time series data and by incorporating Thiru into Geipel and Wren can make models more accurate and operate faster [Thiru ¶22].
As to independent claim 12, Geipel teaches a computer program product comprising a non-transitory computer-readable medium including programmed instructions, the instructions causing a computer to execute: [device, code and memory ¶46, ¶56]
a probability model corresponding to each dimension of each of the plurality of multidimensional probability distribution models being a model that models probabilities of a plurality of values, at a corresponding time, of time series data whose data length is L, the data length of the plurality of pieces of partial time series data being L, [Models assign probability to time series data at interval (length)¶37 "This first and second probabilistic model 225, 235 may each be provided in the form of a function which, for a certain data point or time interval of the time series data 20, outputs a probability of observing this data point time interval. "]
the plurality of pieces of partial time series data being contained in target time series data to be a target of diagnosis; and [device describes time series data dynamics (its diagnosed) in windows (targets) ¶8 "model describing dynamics of the time series data inside the first time window"]
determine a plurality of pieces of matching information including positions of the plurality of pieces of partial time series data in the target time series data, [determines matches ¶8 " determine a first part of the time series data that is estimated to match the first probabilistic model and a second part of the time series data that is estimated to match the second probabilistic model"]
Geipel does not specifically teach calculate first similarities between a plurality of multidimensional probability distribution models and a first probabilistic model whose first similarity with respect to the plurality of pieces of partial time series data at the positions is maximum, and the first similarity to the first probabilistic model.
However, Wren teaches calculate first similarities between a plurality of multidimensional probability distribution models and [measures similarities (likelihood) between model observations Col. 3 ln. 27-38 "if two observations are similar, then the two models constructed and trained for the two observations are similar. That is, the model for one observation generates the other observation with a high likelihood "]
a first probabilistic model whose first similarity with respect to the plurality of pieces of partial time series data at the positions is maximum, and the first similarity to the first probabilistic model. [maximum likelihood (larger) used to determine models Fig. 1 200 among those in the tree 100 Col. 2 ln. 45-52 "Each data sequence is assigned to only a single subtree, specifically, the subtree associated with the single-path model that generates the data sequence with a maximum likelihood. In this way, the single-path models define the patterns in temporal data sequences. The method used to learn the path models from the temporal data sequences is described in detail below."]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the models disclosed by Geipel by incorporating the calculate first similarities between a plurality of multidimensional probability distribution models and a first probabilistic model whose first similarity with respect to the plurality of pieces of partial time series data at the positions is maximum, and the first similarity to the first probabilistic model disclosed by Wren because both techniques address the same field of time series data and by incorporating Wren into Geipel provides better discovering of temporal pattern in data using less time and resources [Wren Col. 1 ln. 42-53]
Geipel and Wren do not specifically teach a plurality of pieces of partial time series data whose data length is L, a number of dimensions of each of the plurality of multidimensional probability distribution models being L,
However, Thiru teaches a plurality of pieces of partial time series data whose data length is L, a number of dimensions of each of the plurality of multidimensional probability distribution models being L, [model dimensions according to components (length) of time-series data ¶38 "anomaly detection envelope or volume having a dimension corresponding to the number of correlated time series data components and/or features"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the models disclosed by Geipel and Wren by incorporating the plurality of pieces of partial time series data whose data length is L, a number of dimensions of each of the plurality of multidimensional probability distribution models being L, disclosed by Thiru because all techniques address the same field of time series data and by incorporating Thiru into Geipel and Wren can make models more accurate and operate faster [Thiru ¶22].
As to dependent claim 14, the rejection of claim 1 is incorporated, Geipel, Wren and Thiru further teach wherein the target time series data includes at least one of: sensor data acquired by one or more sensors that detect physical quantities, and biological signal data that includes at least one of an ElectroCardioGram (ECG) and and ElectroEncephaloGram (EEG). [Geipel sensor data for a pump ¶31], [Wren sensors sequences from building data Col 1 ln. 14-22 "occupants of a building including sensors generate temporal patterns as they move from place to place"]
Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Geipel in view of Wren and Thiru, as applied to the rejection of claim 1 above, and further in view of Vasseur et al. (US 9160760 B2 hereinafter Vasseur)
As to dependent claim 2, the combination of Geipel, Wren and Thiru teach all the limitations of claim 1 that is incorporated.
Geipel, Wren and Thiru do not specifically teach wherein each of the plurality of multidimensional probability distribution models is defined using a mean and covariance.
However, Vasseur teaches wherein each of the plurality of multidimensional probability distribution models is defined using a mean and covariance. [gaussian model with mean and variance Col. 12-13 ln. 57-9 "model reduces to computing of a set of simple statistics (e.g., mean, variance, and covariance)"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify data modelling disclosed by Geipel, Wren and Thiru by incorporating the wherein each of the plurality of multidimensional probability distribution models is defined using a mean and covariance disclosed by Vasseur because all techniques address the same field of managing data and by incorporating Vasseur into Geipel, Wren and Thiru provide more efficient algorithms to handle more data than classic techniques [Vasseur Col. 7 ln. 5-14]
Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable over Geipel in view of Wren and Thiru and Vasseur. as applied to the rejection of claim 2 above, and further in view of Brodersen. et al. (US 20160062950 A1 hereinafter Brodersen)
As to dependent claim 3, the combination of Geipel, Wren, Thiru and Vasseur teach all the limitations of claim 2 that is incorporated.
Geipel, Wren, Thiru and Vasseur do not specifically teach use the mean and the covariance to obtain a normal range where time series data is assumed to be normal, and output information including the normal range.
However, Brodersen teaches use the mean and the covariance to obtain a normal range where time series data is assumed to be normal, and [determines a range of expected values based on mean ¶70 "time-series model will generate a mean and a standard deviation for each data point. The percentage value or the standard deviation multiplier may be used with the mean and the standard deviation to determine whether the data point lies outsides the range of expected values"], [¶107 "Gaussian distribution, with a mean of the previous local level plus the previous local trend 610, with a variance of diffusion variance 605a"]
output information including the normal range. [displays analysis and time series data ¶22]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify data modelling disclosed by Geipel, Wren, Thiru and Vasseur by incorporating the use the mean and the variance to obtain a normal range where time series data is assumed to be normal, and output information including the normal range disclosed by Brodersen because all techniques address the same field of managing data and by incorporating Brodersen into Geipel, Wren, Thiru and Vasseur ensures better detection of events or anomalies in time-series while avoiding overfitting [Brodersen ¶2]
Claims 6-7 is rejected under 35 U.S.C. 103 as being unpatentable over Geipel in view of Wren and Thiru, as applied to the rejection of claim 1 above, and further in view of Calmon et al. (US 11361197 B2 hereinafter Calmon)
As to dependent claim 6, the combination of Geipel, Wren and Thiru teach all the limitations of claim 1 that is incorporated.
Geipel, Wren and Thiru do not specifically teach wherein the one or more hardware processors is configured to detect a state of the target time series data by using the first similarities included in the plurality of pieces of matching information.
However, Calmon teaches wherein the one or more hardware processors is configured to detect a state of the target time series data by using the first similarities included in the plurality of pieces of matching information. [detects states and uses likelihood (similarities) of data samples Col. 1 ln. 31-51 "detected state; obtaining a likelihood (such as a log likelihood) that each of the data samples belongs to the corresponding detected state;"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify data modelling disclosed by Geipel, Wren and Thiru by incorporating the wherein the one or more hardware processors is configured to detect a state of the target time series data by using the first similarities included in the plurality of pieces of matching information disclosed by Calmon because all techniques address the same field of managing data and by incorporating Calmon into Geipel, Wren and Thiru enhances models for proper detection of anomalies according to states of interests avoiding malfunctions and sub-optimal allocation of resources [Calmon Col. 1 ln. 12-28]
As to dependent claim 7, the rejection of claim 6 is incorporated, Geipel, Wren, Thiru and Calmon further teach wherein the one or more hardware processors detect that there is an anomaly in the target time series data, when a minimum value of the first similarities included in the plurality of pieces of matching information is smaller than a threshold. [Calmon detects anomaly in temporal data with thresholds (min) Col. 1 ln. 31-51 "an anomaly detection model that, given the distribution of likelihoods and one or more anomaly thresholds, generates a quality score for each of the one or more anomaly thresholds; and selecting at least one anomaly threshold based on the quality score, wherein the trained anomaly detection model is applied to detect anomalies in new temporal data samples using the selected at least one anomaly threshold"]
Claim 13 is rejected under 35 U.S.C. 103 as being unpatentable over Geipel in view of Wren and Thiru, as applied to the rejection of claim 1 above, and further in view of KAWATSU et al. (US 20210348985 A1 hereinafter Kawatsu)
As to dependent claim 13, the combination of Geipel, Wren and Thiru teach all the limitations of claim 1 that is incorporated.
Geipel, Wren and Thiru do not specifically teach wherein each of the plurality of values is a value of a single variable that is common to a plurality of probability models each corresponding to one of dimensions of each of the plurality of multidimensional probability distribution models.
However, Kawatsu teaches wherein each of the plurality of values is a value of a single variable that is common to a plurality of probability models each corresponding to one of dimensions of each of the plurality of multidimensional probability distribution models. [single sensor information (variable) used as multivariate data which can be used for models ¶47 "reconstruct single sensor information to multivariate data to perform the PCA process"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify data modelling disclosed by Geipel, Wren and Thiru by incorporating the wherein each of the plurality of values is a value of a single variable that is common to a plurality of probability models each corresponding to one of dimensions of each of the plurality of multidimensional probability distribution models. disclosed by Kawatsu because all techniques address the same field of managing data and by incorporating Kawatsu into Geipel, Wren and Thiru enables detections with less data and more accuracy on complex model [Kawatsu ¶70-71]
Response to Arguments
Applicant's arguments filed 03/16/2026, with respect to 101, these rejections have been withdrawn.
Applicant's arguments filed 03/16/2026. In the remark, applicant argues that:
(1) Tomita and Wren fail to teach “calculate first similarities between a plurality of multidimensional probability distribution models and a plurality of pieces of partial time series data whose data length is L, a number of dimensions of each of the plurality of multidimensional probability distribution models being L, a probability model corresponding to each dimension of each of the plurality of multidimensional probability distribution models being a model that models probabilities of a plurality of values, at a corresponding time, of time series data whose data length is L, the data length of the plurality of pieces of partial time series data being L, ” as recited by amended claim 1.
As to point (1), Applicant’s arguments with respect to claim 1 have been considered but are moot in view of a new ground of rejection as set forth above of Geipel in view of Wren and Thiru.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Applicant is required under 37 C.F.R. § 1.111(c) to consider these references fully when responding to this action.
Kawatsu et al. (US 11692910 B2) teaches time-series univariate information into a set of vectors having a measured value of a near point called a partial time series as an element using a sliding window (Col. 7-8 ln. 62-14)
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any extension fee pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BEAU SPRATT whose telephone number is (571)272-9919. The examiner can normally be reached M-F 8:30-5 PST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Jennifer Welch can be reached on 5712127212. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/BEAU D SPRATT/ Primary Examiner, Art Unit 2143
Prosecution Insights